[U-Boot] [PATCH v3 1/4] blk: Use macros for block device vendor/product/rev string size
So far these are using magic numbers. Replace them with macros.
Signed-off-by: Bin Meng bmeng.cn@gmail.com Reviewed-by: Simon Glass sjg@chromium.org
---
Changes in v3: - Change BLK_XXX_SIZE to not count the ending NULL
Changes in v2: None
include/blk.h | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-)
diff --git a/include/blk.h b/include/blk.h index a106f9c..27abfdd 100644 --- a/include/blk.h +++ b/include/blk.h @@ -36,6 +36,10 @@ enum if_type { IF_TYPE_COUNT, /* Number of interface types */ };
+#define BLK_VEN_SIZE 40 +#define BLK_PRD_SIZE 20 +#define BLK_REV_SIZE 8 + /* * With driver model (CONFIG_BLK) this is uclass platform data, accessible * with dev_get_uclass_platdata(dev) @@ -60,9 +64,9 @@ struct blk_desc { lbaint_t lba; /* number of blocks */ unsigned long blksz; /* block size */ int log2blksz; /* for convenience: log2(blksz) */ - char vendor[40+1]; /* IDE model, SCSI Vendor */ - char product[20+1]; /* IDE Serial no, SCSI product */ - char revision[8+1]; /* firmware revision */ + char vendor[BLK_VEN_SIZE + 1]; /* device vendor string */ + char product[BLK_PRD_SIZE + 1]; /* device product number */ + char revision[BLK_REV_SIZE + 1]; /* firmware revision */ #if CONFIG_IS_ENABLED(BLK) /* * For now we have a few functions which take struct blk_desc as a
This converts the IDE driver to driver model so that block read and write are fully functional.
Fixes: b7c6baef ("x86: Convert MMC to driver model") Reported-by: Heinrich Schuchardt xypron.glpk@gmx.de Signed-off-by: Bin Meng bmeng.cn@gmail.com
Reviewed-by: Simon Glass sjg@chromium.org Tested-by: Heinrich Schuchardt xypron.glpk@gmx.de ---
Changes in v3: - Updated BLK_XXX_SIZE usage
Changes in v2: - Fixed 'fatls ide 0' issue
drivers/block/blk-uclass.c | 2 +- drivers/block/ide.c | 70 ++++++++++++++++++++++++++++++++++++++++++++++ include/dm/uclass-id.h | 1 + 3 files changed, 72 insertions(+), 1 deletion(-)
diff --git a/drivers/block/blk-uclass.c b/drivers/block/blk-uclass.c index e5f00dc..8e58580 100644 --- a/drivers/block/blk-uclass.c +++ b/drivers/block/blk-uclass.c @@ -26,7 +26,7 @@ static const char *if_typename_str[IF_TYPE_COUNT] = { };
static enum uclass_id if_type_uclass_id[IF_TYPE_COUNT] = { - [IF_TYPE_IDE] = UCLASS_INVALID, + [IF_TYPE_IDE] = UCLASS_IDE, [IF_TYPE_SCSI] = UCLASS_SCSI, [IF_TYPE_ATAPI] = UCLASS_INVALID, [IF_TYPE_USB] = UCLASS_MASS_STORAGE, diff --git a/drivers/block/ide.c b/drivers/block/ide.c index ce51153..1b5476e 100644 --- a/drivers/block/ide.c +++ b/drivers/block/ide.c @@ -827,12 +827,20 @@ void ide_init(void) ide_ident(&ide_dev_desc[i]); dev_print(&ide_dev_desc[i]);
+#ifndef CONFIG_BLK if ((ide_dev_desc[i].lba > 0) && (ide_dev_desc[i].blksz > 0)) { /* initialize partition type */ part_init(&ide_dev_desc[i]); } +#endif } WATCHDOG_RESET(); + +#ifdef CONFIG_BLK + struct udevice *dev; + + uclass_first_device(UCLASS_IDE, &dev); +#endif }
/* We only need to swap data if we are running on a big endian cpu. */ @@ -1147,6 +1155,23 @@ int ide_device_present(int dev) #endif
#ifdef CONFIG_BLK +static int ide_blk_probe(struct udevice *udev) +{ + struct blk_desc *desc = dev_get_uclass_platdata(udev); + + /* fill in device vendor/product/rev strings */ + strncpy(desc->vendor, ide_dev_desc[desc->devnum].vendor, + BLK_VEN_SIZE + 1); + strncpy(desc->product, ide_dev_desc[desc->devnum].product, + BLK_PRD_SIZE + 1); + strncpy(desc->revision, ide_dev_desc[desc->devnum].revision, + BLK_REV_SIZE + 1); + + part_init(desc); + + return 0; +} + static const struct blk_ops ide_blk_ops = { .read = ide_read, .write = ide_write, @@ -1156,6 +1181,51 @@ U_BOOT_DRIVER(ide_blk) = { .name = "ide_blk", .id = UCLASS_BLK, .ops = &ide_blk_ops, + .probe = ide_blk_probe, +}; + +static int ide_probe(struct udevice *udev) +{ + struct udevice *blk_dev; + char name[20]; + int blksz; + lbaint_t size; + int i; + int ret; + + for (i = 0; i < CONFIG_SYS_IDE_MAXDEVICE; i++) { + if (ide_dev_desc[i].type != DEV_TYPE_UNKNOWN) { + sprintf(name, "blk#%d", i); + + blksz = ide_dev_desc[i].blksz; + size = blksz * ide_dev_desc[i].lba; + ret = blk_create_devicef(udev, "ide_blk", name, + IF_TYPE_IDE, i, + blksz, size, &blk_dev); + if (ret) + return ret; + } + } + + return 0; +} + +U_BOOT_DRIVER(ide) = { + .name = "ide", + .id = UCLASS_IDE, + .probe = ide_probe, +}; + +struct pci_device_id ide_supported[] = { + { PCI_DEVICE_CLASS(PCI_CLASS_STORAGE_IDE << 8, 0xffff00) }, + { } +}; + +U_BOOT_PCI_DEVICE(ide, ide_supported); + +UCLASS_DRIVER(ide) = { + .name = "ide", + .id = UCLASS_IDE, }; #else U_BOOT_LEGACY_BLK(ide) = { diff --git a/include/dm/uclass-id.h b/include/dm/uclass-id.h index 1a50199..3fc2083 100644 --- a/include/dm/uclass-id.h +++ b/include/dm/uclass-id.h @@ -41,6 +41,7 @@ enum uclass_id { UCLASS_I2C_EEPROM, /* I2C EEPROM device */ UCLASS_I2C_GENERIC, /* Generic I2C device */ UCLASS_I2C_MUX, /* I2C multiplexer */ + UCLASS_IDE, /* IDE device */ UCLASS_IRQ, /* Interrupt controller */ UCLASS_KEYBOARD, /* Keyboard input device */ UCLASS_LED, /* Light-emitting diode (LED) */
Hi Tom,
On Wed, Sep 6, 2017 at 3:30 PM, Bin Meng bmeng.cn@gmail.com wrote:
This converts the IDE driver to driver model so that block read and write are fully functional.
Fixes: b7c6baef ("x86: Convert MMC to driver model") Reported-by: Heinrich Schuchardt xypron.glpk@gmx.de Signed-off-by: Bin Meng bmeng.cn@gmail.com
Reviewed-by: Simon Glass sjg@chromium.org Tested-by: Heinrich Schuchardt xypron.glpk@gmx.de
Changes in v3:
- Updated BLK_XXX_SIZE usage
Changes in v2:
- Fixed 'fatls ide 0' issue
drivers/block/blk-uclass.c | 2 +- drivers/block/ide.c | 70 ++++++++++++++++++++++++++++++++++++++++++++++ include/dm/uclass-id.h | 1 + 3 files changed, 72 insertions(+), 1 deletion(-)
Will you take this series in 2017.09?
Regards, Bin
On Sat, Sep 09, 2017 at 08:30:44AM +0800, Bin Meng wrote:
Hi Tom,
On Wed, Sep 6, 2017 at 3:30 PM, Bin Meng bmeng.cn@gmail.com wrote:
This converts the IDE driver to driver model so that block read and write are fully functional.
Fixes: b7c6baef ("x86: Convert MMC to driver model") Reported-by: Heinrich Schuchardt xypron.glpk@gmx.de Signed-off-by: Bin Meng bmeng.cn@gmail.com
Reviewed-by: Simon Glass sjg@chromium.org Tested-by: Heinrich Schuchardt xypron.glpk@gmx.de
Changes in v3:
- Updated BLK_XXX_SIZE usage
Changes in v2:
- Fixed 'fatls ide 0' issue
drivers/block/blk-uclass.c | 2 +- drivers/block/ide.c | 70 ++++++++++++++++++++++++++++++++++++++++++++++ include/dm/uclass-id.h | 1 + 3 files changed, 72 insertions(+), 1 deletion(-)
Will you take this series in 2017.09?
I would swear that we declare vendor/etc as size + 1 so that we can ensure that size+1==NULL and only write up to size into the field. But the rest of this patch strncpy's to size+1. That feels wrong, but I'm not totally sure. Are you? Thanks!
Hi Tom,
On Sat, Sep 9, 2017 at 11:25 AM, Tom Rini trini@konsulko.com wrote:
On Sat, Sep 09, 2017 at 08:30:44AM +0800, Bin Meng wrote:
Hi Tom,
On Wed, Sep 6, 2017 at 3:30 PM, Bin Meng bmeng.cn@gmail.com wrote:
This converts the IDE driver to driver model so that block read and write are fully functional.
Fixes: b7c6baef ("x86: Convert MMC to driver model") Reported-by: Heinrich Schuchardt xypron.glpk@gmx.de Signed-off-by: Bin Meng bmeng.cn@gmail.com
Reviewed-by: Simon Glass sjg@chromium.org Tested-by: Heinrich Schuchardt xypron.glpk@gmx.de
Changes in v3:
- Updated BLK_XXX_SIZE usage
Changes in v2:
- Fixed 'fatls ide 0' issue
drivers/block/blk-uclass.c | 2 +- drivers/block/ide.c | 70 ++++++++++++++++++++++++++++++++++++++++++++++ include/dm/uclass-id.h | 1 + 3 files changed, 72 insertions(+), 1 deletion(-)
Will you take this series in 2017.09?
I would swear that we declare vendor/etc as size + 1 so that we can ensure that size+1==NULL and only write up to size into the field. But the rest of this patch strncpy's to size+1. That feels wrong, but I'm not totally sure. Are you? Thanks!
OK, v4 was just sent. Please take a look and hope it makes it before the release.
Regards, Bin
When there is no CDROM inserted, the block size is zero hence there is no need to create a BLK device for it.
Signed-off-by: Bin Meng bmeng.cn@gmail.com Reviewed-by: Simon Glass sjg@chromium.org ---
Changes in v3: None Changes in v2: None
drivers/block/ide.c | 7 +++++++ 1 file changed, 7 insertions(+)
diff --git a/drivers/block/ide.c b/drivers/block/ide.c index 1b5476e..0631fae 100644 --- a/drivers/block/ide.c +++ b/drivers/block/ide.c @@ -1199,6 +1199,13 @@ static int ide_probe(struct udevice *udev)
blksz = ide_dev_desc[i].blksz; size = blksz * ide_dev_desc[i].lba; + + /* + * With CDROM, if there is no CD inserted, blksz will + * be zero, don't bother to create IDE block device. + */ + if (!blksz) + continue; ret = blk_create_devicef(udev, "ide_blk", name, IF_TYPE_IDE, i, blksz, size, &blk_dev);
At present the IDE device number is initialized to -1, which means we cannot type "ide read" command before setting the device number via "ide device #".
For convenience, let's set the first device as the default one.
Signed-off-by: Bin Meng bmeng.cn@gmail.com Reviewed-by: Simon Glass sjg@chromium.org
---
Changes in v3: None Changes in v2: None
cmd/ide.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/cmd/ide.c b/cmd/ide.c index e3c3242..bdb5980 100644 --- a/cmd/ide.c +++ b/cmd/ide.c @@ -30,7 +30,7 @@ #endif
/* Current I/O Device */ -static int curr_device = -1; +static int curr_device;
int do_ide(cmd_tbl_t *cmdtp, int flag, int argc, char *const argv[]) {
participants (2)
-
Bin Meng -
Tom Rini